Gemini 2.5 Flash (Medium): benchmark results
Provider: Google. Released 2025-06-17. Access: API.
Aggregate Rating 1596 ± 24, rank #743 of 2170 rated models, from 42 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| IKP - T1 | 100 | Accuracy (%) | 86.1 |
| CommunityFact (Closed-Input) - Finance (Portuguese) | 69 | Macro-F1 (%) over True and False claims on the temporally he | 85.7 |
| IKP - T7 | 14.53 | Accuracy (%) | 83.1 |
| IKP - T6 | 33.16 | Accuracy (%) | 82.9 |
| CommunityFact (Closed-Input) - Finance (Japanese) | 78.48 | Macro-F1 (%) over True and False claims on the temporally he | 82.1 |
| TuRTLe - Icarus Functionality | 70.49 | Average Score (%) | 81.8 |
| TuRTLe - Verilator Functionality | 70.94 | Average Score (%) | 81.8 |
| TuRTLe Code Completion (Icarus Verilog) | 69.84 | Aggregated Score (imported) | 79.1 |
| TuRTLe Code Completion (Verilator) | 70.19 | Aggregated Score (imported) | 79.1 |
| CommunityFact (Closed-Input) - Finance (French) | 63.51 | Macro-F1 (%) over True and False claims on the temporally he | 78.6 |
| CommunityFact (Closed-Input) - Politics (English) | 67.45 | Macro-F1 (%) over True and False claims on the temporally he | 78.6 |
| CommunityFact (Closed-Input) - Politics (French) | 65.71 | Macro-F1 (%) over True and False claims on the temporally he | 78.6 |
Interactive version: theaggregate.ai/model?slug=gemini-2-5-flash-medium · How It Works · Data refreshed daily, snapshot 2026-11-05.